integrin a3b1 can function to promote spontaneous ... · signal transduction integrin a3b1 can...

Signal Transduction

Integrin a3b1 Can Function to Promote SpontaneousMetastasis and Lung Colonization of Invasive BreastCarcinoma

Bo Zhou1, Katherine N. Gibson-Corley2, Mary E. Herndon1, Yihan Sun1, Elisabeth Gustafson-Wagner1,Melissa Teoh-Fitzgerald4, Frederick E. Domann2,4,5, Michael D. Henry2,3,5, and Christopher S. Stipp1,3,5

AbstractSignificant evidence implicatesa3b1 integrin in promoting breast cancer tumorigenesis andmetastasis-associated

cell behaviors in vitro and in vivo.However, the extent towhicha3b1 is actually required for breast cancermetastasisremains to be determined. We used RNA interference to silence a3 integrin expression by approximately 70% in4T1murine mammary carcinoma cells, a model of aggressive, metastatic breast cancer. Loss of a3 integrin reducedadhesion, spreading, and proliferation on laminin isoforms, and modestly reduced the growth of orthotopicallyimplanted cells. However, spontaneous metastasis to lung was strikingly curtailed. Experimental lung colonizationafter tail vein injection revealed a similar loss of metastatic capacity for the a3-silenced (a3si) cells, suggesting thatcritical, a3-dependent events at the metastatic site could account for much of a3b10s contribution to metastasis inthismodel. Reexpressinga3 in thea3si cells reversed the loss ofmetastatic capacity, and silencing another target, thesmall GTPase RhoC, had no effect, supporting the specificity of the effect of silencing a3. Parental, a3si, anda3-rescued cells, all secreted abundant laminin a5 (LAMA5), an a3b1 integrin ligand, suggesting that loss of a3integrin might disrupt an autocrine loop that could function to sustain metastatic growth. Analysis of human breastcancer cases revealed reduced survival in cases where a3 integrin and LAMA5 are both overexpressed.

Implications: a3 integrin or downstream effectors may be potential therapeutic targets in disseminated breastcancers, especially when laminin a5 or other a3 integrin ligands are also over-expressed.Mol Cancer Res; 12(1);143–54. �2013 AACR.

IntroductionNormal mammary epithelia are surrounded by the base-

ment membrane, an extracellular matrix rich in laminin

isoforms, including laminin-332 (LM-332; LAMA3/LAMB3/LAMC2) and laminin-511 (LM-511; LAMA5/LAMB1/LAMC1). Early studies revealed that mammarycarcinoma cells can co-opt LM-332 to promote anchor-age-independent growth and survival (1, 2), and that LM-332 can potently promote breast cancer cell migration (3).Although early studies of clinical breast cancer specimenssuggested that LM-332 expression is often lost duringprogression from ductal carcinoma in situ to invasive breastcancer (4–7), LM-332 may be retained in certain breastcancers, such as metaplastic breast carcinoma (8, 9), and in asignificant fraction of triple-negative, basal-like breast cancers(10). Moreover, LM-332 may be upregulated in the reactivestroma adjacent to invasive ductal carcinomas (11). Inaddition, compared with LM-332, LM-511 may more oftenbe retained in advanced breast cancer (12–14), reviewed inref. (15). LM-511 is also abundant in adult bonemarrow (16,17) and lung stroma (18) and thus may be a relevantextracellular ligand for tumor cells at metastatic sites.Breast carcinoma cells engage laminin isoforms via integ-

rins a3b1 (ITGA3/ITGB1) and a6b4 (ITGA6/ITGB4).Expression of b4 integrin and a coregulated gene set corre-lates with a more aggressive malignant phenotype in breastcancer (19, 20), and numerous functional studies haveestablished a role for a6b4 integrin in promoting cancer

Authors' Affiliations: Departments of 1Biology, 2Pathology, 3MolecularPhysiology & Biophysics; 4Free Radical and Radiation Biology Program,Department of Radiation Oncology; and 5Holden Comprehensive CancerCenter, University of Iowa, Iowa City, Iowa

Note: Supplementary data for this article are available at Molecular CancerResearch Online (http://mcr.aacrjournals.org/).

K.N. Gibson-Corley and M.E. Herndon contributed equally to this work.

Current address for B. Zhou: Fels Institute for Cancer Research andMolecular Biology, TempleUniversity School ofMedicine, Philadelphia, PA.

Current address for Y. Sun, University of Michigan College of Pharmacy,Ann Arbor, MI.

Current address for E. Gustafson-Wagner, Integrated DNA Technologies,Coralville, IA.

Current address for M. Teoh-Fitzgerald, Department of Biochemistry andMolecular Biology, University of Nebraska Medical Center, Omaha, NE.

CorrespondingAuthor:Christopher S. Stipp, University of Iowa; BBE 236;Iowa City, IA 52242. Phone: 319-335-0192; Fax: 319-335-1069; E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-13-0184

�2013 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org 143

on June 25, 2020. © 2014 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst September 3, 2013; DOI: 10.1158/1541-7786.MCR-13-0184

http://mcr.aacrjournals.org/

cell survival, anchorage-independent growth, invasion, andmetastasis (reviewed in ref. 21–23). The tumor-promotingactivities of integrin a6b4 require the signaling functions ofthe unusually large b4 integrin cytoplasmic tail and caninvolve activation of RAC signaling toward NF-kB (2),phosphoinositide 3-kinase association with insulin receptorsubstrate-1/2 and signaling toward AKT and RAC (24, 25),regulation of cyclic AMP levels and the interplay betweenRHO, RAC, and protein kinase A (PRKCA) activity (26–28), stimulation of autocrine VEGF signaling (29), cross-talkwith growth factor receptors (30–32), and phosphatase SHP2(PTPN11) signaling toward multiple downstream effectors,including the FYN tyrosine kinase (30, 33, 34). Some a6b4oncogenic signaling functions may be independent of ligandbinding (31), but others require ligand engagement (35).Significant evidence also implicates integrin a3b1 as a

regulator of breast cancer progression. However, the picturethat has emerged of a3b1 functions in breast cancer isperhaps less clear than that of a6b4 integrin. Some earlystudies described an association between the loss of a3integrin in primary breast cancer specimens and the presenceof lymph node metastases (36, 37). However, other studiesrevealed that a3b1 can contribute to breast carcinoma celladhesion to lymphnode stroma in cryostat sections (38) or tocortical bone disks, in an in vitromodel of events relevant tobone metastasis (39). In one study, antibody ligation of a3integrin onMDA-MB-231 breast carcinoma cells enhancedproduction of active matrix metalloproteinase-2 (MMP2),increased protrusive activity in 3D Matrigel, and increasedMatrigel invasion (40). Yet a different group reported thatantibody ligation of a3b1 on the same cell type impairedproduction of MMP9 and reduced Matrigel invasion (41).In favor of the view that a3b1 can contribute to themetastatic behavior of breast cancer cells, antibody ligationofa3b1 reduced (by�30%) the number of MDA-MB-231cells detected in the lungs after injection in a rat tail veinmodel of pulmonary arrest (42). Perhaps the strongestexperimental evidence to date that a3b1 can promote breastcancer progression in vivo comes from Mitchell and collea-gues (43), who showed that RNA interference (RNAi)silencing of a3 in MDA-MB-231 cells suppressed tumorgrowth at both subcutaneous and orthotopic sites. Thus,a3b1 integrin can play an important role in the growth ofprimary breast tumors; however, its role in metastatic dis-semination and growth at secondary sites requires furtherevaluation. Moreover, our recent finding that a3b1 can actas a suppressor of metastatic colonization in a model ofaggressive prostate cancer (44) underscores the need tocarefully examine a3b1 in a variety of metastasis models,to understand the range of a3b1 functions in metastasis.To evaluate a3b1 integrin's potential to contribute to

breast cancer metastasis, we selected the 4T1 murine mam-mary carcinomamodel. In the 4T1model, small numbers oftumor cells implanted orthotopically in immunocompetentmice give rise to rapidly growing primary tumors thatmetastasize spontaneously and form macroscopic coloniesin the lung and other clinically relevant sites. In addition, the4T1 model has been reported to resemble basal-like triple-

negative ductal carcinoma (45), an aggressive molecularsubtype of human breast cancer. Using this model, we nowprovide evidence thata3b1 integrin can play a critical role inspontaneous metastasis of breast carcinoma cells by a mech-anism that may involve autocrine production of the a3b1ligand, laminin-511.

Materials and MethodsAntibodies and extracellular matrix proteinsAntibodies used in this study were rabbit anti-a3 integrin

cytoplasmic tail antibody, A3-CYT (46), rabbit anti-mouselaminin a5 (LAMA5; ref. 47), mouse anti-myc epitope tag,(clone 9E10, Developmental Studies Hybridoma Bank),rabbit anti-RHOC (clone D40E4, Cell Signaling Inc), ham-ster anti-mouse a2 (ITGA2) integrin, (clone HMa2,eBioscience), rat anti-mouse a6 integrin (clone GoH3,eBioscience), and goat anti-mouse Cox-2 (PTGS2; M19,sc-1747, Santa-Cruz Biotechnology). Alexa 488-conjugatedgoat-anti-rabbit, goat-anti-mouse, and goat-anti-rat secondaryantibodies were purchased from Invitrogen. Cy2-conjugatedgoat-anti-hamster was from Jackson ImmunoResearch. Extra-cellular matrix proteins used in this study were laminin-332(purified from SCC25 squamous carcinoma cell-conditionedmedium as previously described; ref. 48), rat tail collagen I(BD Bioscience), and laminin-511 (BioLamina AB).

Cell culture and RNA interference4T1 breast carcinoma cells (American Type Culture

Collection; ATCC) were cultured in RPMI medium.GP2-293 retroviral packaging cells (Clontech) and MDA-MB-231 cells (ATCC) were cultured in Dulbecco's mod-ified Eagle medium. All 4T1 and MDA-MB-231-derivedcell lines were created from early passage cells that had beencultured for less than 6 months after resuscitation. Growthmedia were supplemented with 10% FBS (Valley Biomed-ical), and 2 mmol/L L-glutamine, 100 U/mL penicillin, and100 mg/mL streptomycin (Invitrogen). EGW593.Lu cellsare MDA-MB-231 cells that were recovered from a lungmetastasis after orthotopic implantation of the parental celltype in the mammary fat pad of a female severe combinedimmunodeficient mouse.Retroviral particles were produced by transfection ofGP2-

293 cells with retroviral vectors, 0.45 mm filtering of virus-containing supernatants, and supplementing with 4 mg/mLpolybrene. To facilitate monitoring tumor growth in vivo,4T1 cells were first transduced with a luciferase cDNA inretroviral vector pQCXIN (Clontech), and selected withG418. For RNAi silencing of a3 integrin, double-strandedoligonucleotides encoding short hairpin RNAs (shRNA)targeting the murine a3 integrin mRNA were cloned intoa modified pSIREN-RetroQ retroviral vector (BD Bios-ciences) harboring the hygromcyin resistance maker. Trans-duced cells were selected with hygromycin and tested for a3integrin expression by cell surface labeling and immunopre-cipitation with A3-CYT anti-a3 integrin antibody. Weidentified one effectivea3 shRNA,with a targeting sequenceof 50-GTCCTATCGTCATTGCCATGA-30. To restorea3 expression in the a3-silenced (a3si) cells, we obtained

Zhou et al.

Mol Cancer Res; 12(1) January 2014 Molecular Cancer Research144




a murine a3 integrin cDNA (catalog # MMM1013-9202265, source ID# 6401146, Open Biosystems) andused recombinant PCR to (i) remove the 50 and 30-untrans-lated regions, (ii) introduce 4 silent mutations within theshRNA-targeting sequence, and (iii) add restrictions sites tofacilitate cloning in frame with a 30 Myc epitope tag in thepLXIZ retroviral vector. Stably transduceda3 rescue (a3Rx)cells were selected with zeocin. To silence a3 integrin inhuman EGW593.Lu cells, shRNA constructs were preparedas described above using targeting sequences of 50-TC-ACTCTGCTGGTGGACTATA-30 (a3sh3) and 05-GGA-TGACTGTGAGCGG-ATGAA-30 (a3sh4). Throughoutthe process of creating luciferase-labeled, a3 silenced anda3Rx cells, all cell populations were maintained as stable,polyclonal populations to minimize concerns about clonalvariation. Bioluminescence imaging of 4T1 parental, a3si,and a3Rx cells revealed that all three lines had very similarluminescence of approximately 40 photons/second/cell.

Cell surface labeling and flow cytometry4T1 cells were labeled on ice with sulfo-NHS-LC-biotin

(Thermo-Fisher Pierce), lysed in 1%TritonX-100 detergent(Sigma-Aldrich), and proteins were immunoprecipitatedfrom clarified lysates, as previously described (48). Immu-noprecipitates were resolved by SDS-PAGE, transfered tonitrocellulose, and visualized by blotting with IRDye-800-streptavidin (Rockland Immunochemicals, Inc) and scan-ning with an Odyssey infrared imaging system (LI-CORBiosciences). For flow cytometry, immunostained cells wereanalyzed on a Becton Dickinson FACScan flow cytometer.

Adhesion and spreading assaysFor adhesion assays, wells were coated overnight with the

indicated concentrations of LM-332, with 20 mg/mL col-lagen I or with 10 mg/mL heat-inactivated bovine serumalbumin (BSA; negative control). Wells were rinsed andblocked with 10 mg/mL heat-inactivated BSA. 4T1 cellswere starved overnight in serum-free medium (SFM), and20,000 cells/well were plated in SFM in each of the foursubstrate-coated wells per condition in a 96-well plate. After25 minutes at 37�C, 5% CO2, wells were rinsed three timesand adherent cells were fixed and quantified by staining withcrystal violet, as previously described (48).To assess short-term cell spreading, cells were grown

overnight in SFM, harvested as described above for the adh-esion assay, and plated on acid-washed glass coverslips thathad been coatedwith1mg/mLLM-332or 20mg/mLcollagenI. After 30 minutes, cells were fixed and photographed on aLeica DMIRE2 inverted microscope using a 20X phaseobjective. ImageJ software (49) was used to measure thespread area of at least 98 cells per cell type. Specific spreadingwas calculated by subtracting the mean area of unspread cells(fixed immediately after plating) from the spread cell areasmeasured at the end of the assay, as in ref. (50).

Spontaneous metastasis assayAll animal procedures were conducted according to the

University of Iowa Animal Care andUseCommittee policies

(Iowa City, IA). Female BALB/c mice (NCI-Frederick) wereimplantedwith 5,000 cells in a volume of 50mL in the fourthmammary fat pad. Bioluminescent imaging (BLI) was con-ducted in an IVIS100 imaging system (Caliper Life Sciences)after intraperitoneal injection of luciferin (100 mL of 15 mg/mL solution per 10 g) as described previously (51). Wholebody tumor growth rates were measured as follows: arectangular region of interest was placed around the dorsaland ventral images of each mouse, and total photon flux(photons/sec) was quantified using Living Image softwarev2.50 (Caliper Life Sciences). The dorsal and ventral valueswere summed and mean BLI values for each group wereplotted biweekly. Primary tumor growth was also measuredby caliper, and tumor volumes were calculated using theformula 1/2(L � W2).To quantify spontaneous metastasis to lung, ex vivo BLI

was conducted on lungs harvested at assay endpoint (day 31or day 35). To ensure the best possible uniformity ofmeasurement conditions, mice were sacrificed in groups of2 or 3, and lungs were immediately harvested and imaged exvivo. All lungs were imaged within 20 to 30 minutes aftereuthanasia. 4T1 cells colonizing the lungs were recovered bymincing the lungs with a sterile razor blade and digestingwith 200 U/mL collagenase II (Worthington BiochemicalCorp.) in complete medium for 15 minutes at 37�C.Explanted cells were grown out under G418 selection foranalysis ofa3 integrin expression by cell surface labeling andimmunoprecipitation, as described above. Some lungs werefixed and processed for histology as described for the exper-imental metastasis assay, below.

Experimental metastasis assayFemale BALB/c mice were injected with 5� 104 cells via

lateral tail vein in a volume of 200 mL. BLI and quanti-fication of whole body tumor growth rates were conductedas described above for the spontaneous metastasis assay.Kaplan–Meier analysis of survival was conducted usingPrism 5 (GraphPad Software) on the basis that day 0 wasthe day of tail vein injections and the endpoint was the dayof euthanasia as determined by more than 10% bodyweight loss, hind limb paralysis or fracture, immobility,or a total photon flux more than 1� 108, a value that initialresults indicated reliably predicts death in less than oneweek in this model. Lungs harvested at the endpoint werefixed in 4% paraformaldehyde overnight at 4�C, rinsed andtransferred to 30% ethanol, and stored at 4�C until furtheranalysis.For histology, lungs were routinely processed, paraffin

embedded, and three 4-mm thick sections were taken perlung at 300 mm intervals and stained with hematoxylin andeosin (H&E) for a microscopic tumor count. Tumor countswere defined by counting tumor nodules. Tumor nodulesoften times coalesced and were difficult to decipher. In thesecases, large nodules were counted as only one nodule. Three4� fields of view were counted per slide. Three slides permouse were analyzed with 3 mice per group. Lungs werechosen for histology to reflect the mean value for each groupas measured by BLI.

Integrin a3b1 Promotes Breast Cancer Metastasis

www.aacrjournals.org Mol Cancer Res; 12(1) January 2014 145




Proliferation assayWells were coated with 1 mg/mL LM-332, 1 mg/mL LM-

511, 10 mg/mL collagen I, or left uncoated. Wells wererinsed 3 times with PBS and blocked with 5 mg/mL BSA inRPMI with 40 mmol/L HEPES buffer, pH 7.2. A total of5,000 cells in 200 mL of RPMI plus 2% FBS was plated in 6wells per cell type per condition in a 96-well plate. Tomeasure cells input on day 0, an additional set of 6 wells percell type was plated in collagen-coated wells, and assayed 2hours later. Plates were developed by discarding 100mL fromeach well and adding 100 mL of solution containing RPMIwith 2% FBS and WST-1 reagent (Roche Diagnostics)diluted 1:10. Plates were incubated for 1 hour at 37�C, andabsorbance at 440 nm was measured using a plate reader.

Laminin a5 immunostaining4T1 cells were plated on acid-washed glass coverslips

coated with 10 mg/mL collagen I and cultured overnight.Cells were fixed and stained with anti-mouse LAMA5 rabbitpolyclonal antiserum #8948 (generous gift of Jeffrey Miner,Washington University, St. Louis, MO) followed by Alexa488 goat anti-rabbit, Alexa 594 phalloidin (Invitrogen), and0.5 mg/mL 40, 6-diamidino-2-phenylindole (DAPI). Bright-ness and contrast of fluorescent images were adjusted withidentical parameters for each cell type using the Adjust >Window/Level command in ImageJ, and overlaid using theOverlay command.

Analysis of data from The Cancer Genome AtlasThe Cancer Genome Atlas (TCGA) Breast Invasive Car-

cinoma provisional dataset was queried using the cBioPortalinterface (52, 53). All tumors with mRNA data (RNA SeqV2; 914 samples as of July 2013) were tested for a3 integrin(ITGA3) overexpression (Z-score > 2) revealing overexpres-sion in 6% of the samples. The cases affected were retrievedusing the download tab and used to create a user-defined caselist, which was then queried a second time for upregulationof LAMA5 mRNA (Z-score > 1), revealing upregulation in30% of the cases. Survival data for the LAMA5-upregulatedversus LAMA5 not upregulated cases were downloaded fromthe interface.

ResultsStable silencing of a3 integrin in the 4T1 mammarycarcinomaTo evaluate the role of a3b1 integrin in a model of

spontaneous breast cancer metastasis, we used retroviralRNAi vectors to stably silence a3 integrin in 4T1 murinemammary carcinoma cells. Cell surface labeling followed byimmunoprecipitation revealed a substantial reduction of a3expression in thea3si cells comparedwithwild-type parentalcells (Fig. 1A, lanes 1 and 2). We restored a3 expression inthe a3si cells by introducing a myc epitope-tagged, RNAi-resistant a3 expression vector to create a3Rx cells (Fig. 1A,lane 3). Anti-myc immunoprecipitation confirmed theexpression of the a3Rx construct in the a3Rx cells (Fig.1A, lane 6). Quantification of several independent blotsrevealed that a3 expression was reduced by approximately

Figure 1. Impaired adhesion and spreading of a3 integrin-silenced 4T1carcinoma cells. A, cell surface biotinylated wild-type (WT), a3si, anda3Rx4T1cellswere lysed in 1%Triton X-100detergent anda3b1 integrinwas immunoprecipitated using the A3-CYT polyconal antibody (lanes1–3) or the 9E10 anti-myc epitope antibody (lanes 4–6). The blot wasvisualized with DyLight 800-Neutravidin. Arrows indicate the a3 and b1integrin bands. B, quantification ofmultiple independent blots by LI-CORinfrared fluorescent scanner. Data are presented as% wild-typeexpression � SEM, n ¼ 6 blots. C, wild-type, a3si, and a3Rx 4T1 cellswere plated in wells with different coating concentrations of LM-332.After 25 minutes, nonadherent cells were removed by rinsing, andadherent cells were quantified by crystal violet staining. Compared withwild-type or a3Rx cells, the a3si cells showed significantly reducedadhesion onwells coated 0.5 mg/mL and 1.0 mg/mL LM-332 (�,P < 0.001,ANOVA with Tukey posttest, n ¼ 4 wells/condition). D, wild-type, a3si,and a3Rx 4T1 cells were plated on wells coated with 20 mg/mL collagenI or in BSA-blocked–negative control wells. After 25 minutes, adherentcells were quantified as in (A). E–G, wild-type, a3si, and a3Rx 4T1 cellswere plated on glass coverslips coated with 1 mg/mL LM-332 for 30minutes and then fixed and photographed with differential interferencemicroscopy. H, the spread area of wild-type, a3si, and a3Rx cells wasquantified with ImageJ software, as described in Materials andMethods.The spread area ofa3si cellswas significantly less thanwild-typeora3Rxcells (�, P < 0.01, ANOVA with Tukey posttest, n � 98 cells per cell type).

Zhou et al.





70% in thea3si cells compared with wild-type, and restoredto approximately 80% of wild-type level in the a3Rx cells(Fig. 1B). The expression levels of a6 integrin, a2 integrin,andb1 integrinwere not dramatically altered in thea3si cells(Supplementary Table S1). Cell surface labeling followed bya6 integrin immunoprecipitation revealed that a6 pairswith the b4 subunit in 4T1 cells and confirmed thata6b4 expression was similar in all three cell types (Supple-mentary Fig. S1). The apparent expression level of a6b4appeared modest compared with that of a3b1 in 4T1parental cells. Flow cytometry of a3 itself is not possibledue to lack of a suitable anti-murine a3 antibody directed atthe a3 ectodomain.Compared with wild-type or a3Rx cells, the a3si cells

displayed a significant reduction in adhesion on LM-332(Fig. 1C). All three cell types adhered strongly to the a2b1integrin ligand, collagen I, and showed minimal adhesion tothe BSA-negative control (Fig. 1D). In short-term spreadingassays on LM-332, the a3si cells displayed reduced lam-mellipodium formation compared with wild-type or a3Rxcells (compare Fig. 1Fwith Fig. 1E andG).Quantification ofspread area on LM-332 revealed a reduction for thea3si cellsof approximately 45% (Fig. 1H). Collectively, these dataestablished that a3b1integrin function was specificallyimpaired in the a3si cells and restored in the a3Rx cells.

Reduced primary tumor growth and spontaneousmetastasis of a3 integrin-silenced cellsTo assess the role of a3b1 integrin in breast carcinoma

cells in vivo, we implanted wild-type,a3si, anda3Rx cells inmammary fat pads of immunocompetent BALB/c mice.Monitoring tumor growth by BLI, we observed a reducedapparent tumor burden in mice harboring the a3si cellsthrough day 10 (Fig. 2A and B; BLI color version inSupplementary Fig. S2A). At time points beyond day 10,the a3si cell tumor burden as measured by BLI was indis-tinguishable from wild-type 4T1 cell tumor burden (Fig. 2Aand B). However, caliper measurements of tumor volumerevealed that a3si cell tumors were somewhat smaller thanwild-type or a3Rx tumors throughout the assay, except onthe final day (Fig. 2C). A potential explanation for thediscrepancy between the BLI measurements and the tumorvolume measurements is that large 4T1 cell tumors maydisplay extensive areas of internal necrosis (SupplementaryFig. S3). Thus, the number of viable cells capable ofcontributing to the BLI signal in the larger, wild-type, anda3Rx tumors may constitute only a fraction of the totaltumor mass, as has been reported for a bladder cancer model(54). Additional factors limiting BLI signal in large tumorsmay include increased optical density and reduced perfusionof poorly vascularized tumor cores with luciferin (55).

Figure 2. Growth of primary tumors and total tumor burden. On day 0, 5,000 luciferase-expressing 4T1 wild-type, a3si, and a3Rx cells were implanted in thefourth mammary fat pad of female Balb/C mice. A, BLI of the cells on day 10 (when a3si tumor cell burden appeared reduced compared with controls)and day 31 (whena3si tumor cell burden appeared similar towild-type). B, total apparent tumor burden,measured as log photon flux, for the entire timecourseof the experiment. The a3si tumor cell burden was significantly less than wild-type or a3Rx on days 6 and 10 (�, P < 0.001 vs. wild-type, P < 0.01 vs.a3Rx, ANOVA with Tukey posttest; n ¼ 10 mice per group). The slight reduction in photon flux observed on day 35 was due to the loss of some of themicewith thehighest tumor burdensbetweenday31andday35.C, tumor volumesmeasuredbycaliper. Themean volumeof thea3si tumorswassignificantlyless than the volumesof both thewild-type anda3Rx tumors fromday20 throughday 31 (�,P <0.01onday20, andP <0.05ondays 24–31, ANOVAwith Tukeyposttest, except for day 27, when a3si was significantly different from a3Rx, but not wild-type).






To assess spontaneous metastasis, we measured lungcolonization by ex vivo BLI at the assay endpoint. Comparedwithwild-type ora3Rx cells,a3si 4T1 cell lung colonizationwas dramatically reduced, and quantification revealed anapparent approximately 10-fold reduction in spontaneouslung colonization by the a3si cells (Fig. 3A; BLI image inSupplementary Fig. S2B). Histologic analyses of lung col-onization, both in these spontaneousmetastasis assays and inexperimental metastasis assays after tail vein injection, sup-ported the results obtained fromBLImeasurements (See Fig.5, below, and Supplementary Fig. S4).In further support of the specificity of the effect of

silencing a3 integrin, we established 4T1 lines harboringtwo separate shRNAs targeting the small GTPase, RhoC,which can promote breast cancer metastasis in other systems(56, 57). Despite near total silencing of RhoC, which wasmaintained during in vivo passaging, these RNAi constructshad no impact on the growth of the primary tumor or onspontaneous metastasis to lung (Supplementary Fig. S5).These data indicate that neither transduction with our RNAivector nor instigation of active RNAi within 4T1 cells is byitself sufficient to alter their metastatic capacity. The dataalso indicate that RhoC seems not to be required for 4T1 cellspontaneous metastasis. The lack of impact upon silencingRhoC alone might be due to pleiotropic effects of transcrip-tion factor TWIST1 and its target miR-10b in the 4T1 cellline (58, 59). Twist signaling through miR-10b can coor-dinately upregulate both RhoC and RhoA, among othertargets, to drive breast cancer cell invasion (60).To confirm that a3 silencing was retained in vivo over the

5-week assay,we isolated 3 sublines of each cell type from lungexplants. As shown in Fig. 3B, the loss of a3 was maintainedin the in vivo-deriveda3si lines. Because thea3si cell primarytumors grew more slowly, especially as assessed by tumorvolume, we examined the relationship between primarytumor volume and lung colonization. There was no strong

relationship between the size of the primary tumor and theextent of lung colonization for any of the three cell types (Fig.3C). Together, these data indicated that silencinga3 integrindelayed the growth of the primary tumor, especially at earliertime points, but had an even greater impact on the number oftumor cells that ultimately colonized the lung.

Dramatic reduction in experimental lung metastasis ofa3 integrin-silenced cellsTo begin to determine which steps in the metastatic

cascade might critically depend upon a3b1, we nextexamined experimental lung colonization by cells injectedvia tail vein. Bioluminescence imaging (BLI) immediatelypostinjection confirmed that cells arrested in the lungs(Fig. 4A; BLI color version in Supplementary Fig. S2C).Two weeks postinjection, BLI revealed a dramatic reduc-tion in lung colonization by a3si cells compared withwild-type or a3Rx cells (Fig. 4A). Quantification of BLIimages confirmed that a3si cell colonization was signif-icantly impaired from day 12 onward, with over an orderof magnitude reduction in apparent tumor burden evidentby day 18 (Fig. 4B). In addition, survival of mice bearinga3si tumor cells was significantly enhanced comparedwith mice bearing wild-type or a3Rx cells (Fig. 4C). Inboth the BLI and survival analyses, the a3Rx cells dis-played an intermediate phenotype at some time points,suggesting that a3 function may not be completelyrescued in the a3Rx cells with regards to experimentallung colonization.To confirm the results of the BLI analysis, we also con-

ducted gross and histologic analysis of lungs recovered at theendpoint of the assay. Compared with lungs from miceharboring wild-type or a3Rx 4T1 cells, far fewer metastaticnodules were grossly evident on the surface of lungs harboringa3si 4T1 cells (Fig. 5A). Quantification of tumor nodules inH&E-stained sections confirmed that a3si cells formed

Figure 3. Spontaneous metastasis is significantly impaired in a3 integrin-silenced 4T1 cells. A, lung colonization as quantified in ex vivo BLI images(see Supplementary Fig. S2B for representative images). The a3si 4T1 cell colonization was significantly reduced compared with wild-type or a3Rx cellcolonization (�, P < 0.05 vs. wild-type, P < 0.001 vs. a3Rx, ANOVA with Tukey posttest; n¼ 10–14 mice per group; bars indicate mean� SEM). B, wild-type,a3si, and a3Rx 4T1 cells were cell surface biotinylated, and a3b1 integrin was recovered by immunoprecipitation from Triton X-100 lysates, as in Fig. 1. Foreach set, the lane marked "parental" corresponds to the original cell line implanted into mammary fat pad at the beginning of the assay, and the lanesnumbered 1 to 3 correspond to sublines recovered from lung explants at the end of the assay. Quantification of band intensities by LI-COR blotimager is indicated under each lane (in arbitrary intensity units). C, the relationship between primary tumor volume and lung colonization measured byBLI is graphed for each cell type. Values for R2 and P from Pearson correlation tests are shown for each graph. In no case was there a significant correlationbetween primary tumor volume and lung colonization.

Zhou et al.





significantly fewer nodules than either wild-type or a3Rxcells (Fig. 5B–E). Again, the a3Rx cells displayed an inter-mediate phenotype, consistent with the BLI data. We alsoconducted a histologic analysis of archived lungs ofmice fromthe spontaneous metastasis assay in Figs. 2 and 3. Althoughthe data were more highly variable due to the fact that fewerintact lungs were available, the histologic results were inagreement with the BLI analysis, witha3si 4T1 cells formingmany fewer tumor nodules (Supplementary Fig. S4). Aninteresting feature noted in Supplementary Fig. S4 (inset) isthe large number of neutrophils throughout the pulmonaryvasculature in these mice by the end of the spontaneousmetastasis assay, consistent with an earlier report (61). Thelarge number of intravascular neutrophils present suggeststhat regulationof tumor cell–neutrophil interactionsmaybe afactor to consider in future studies of a3 integrin function inthe 4T1 metastasis model. In sum, these data revealed that,whatever is its contribution to early invasive events at theprimary site,a3b1 integrin is also likely to strongly influencethe ability of breast carcinoma cells to colonize the lung oncethey have disseminated from the primary tumor.

a3b1-silenced 4T1 cells display diminishedproliferation on laminin isoforms and secrete LM-511,an endogenous a3b1 ligandIntegrin a3b1 may promote prometastatic breast cancer

cell behaviors via multiple downstream pathways including

regulation of COX-2, VEGF secretion, and MMP produc-tion (62). However, we were unable to detect significantdifferences in COX-2, VEGF, or MMP production in our4T1 cell lines (Supplementary Fig. S6A–S6C). Transen-dothelial migration assays of our 4T1 cell lines suggested amodest decrease inmigration for thea3si cells, but the effectdid not reach statistical significance (Supplementary Fig.S6D). In short-term proliferation assays, the a3si cellsdisplayed reduced proliferation not only on LM-332 andLM-511, but also on collagen I or uncoated plastic (Fig. 6).The reduced proliferation of the a3si cells irrespective of theexogenous ligand supplied suggested the possibility that anendogenous a3 integrin ligand may be contributing toproliferation in our assay.Pouliot and Kusuma have reported that a bone-metastatic

subclone of 4T1 cells expresses LM-511, which may be animportant regulator of metastastic colonization in this sys-tem (14, 63, 64). Immunostaining cultures of our wild-type,a3si, and a3Rx cells for LM-511 revealed that all three of

Figure 5. Gross and histologic analysis of experimental lung metastasis.A, lungs recovered 18 to 28 days after injection from mice bearing wild-type, a3si, or a3Rx 4T1 cells were paraformaldehyde fixed andphotographedusingadissectingmicroscopebeforeparaffinembedding.Pulmonary metastases appear as pale tan nodules on the surfaces oflungs (arrows point to examples). Fewer nodules were evident on lungsfrommicewitha3si cells. B, quantification of tumor nodulesper 4� field inH&E-stained sections from paraffin-embedded lungs. The a3si 4T1 cellsformed fewer nodules than either wild-type or a3Rx 4T1 cells. The wild-type cells formed more nodules than the a3Rx cells, indicating a partialrestoration of a3 function for the a3Rx cells in this assay. (�, P < 0.01,ANOVA with Tukey posttest, n¼ 3 mice per group, 3 slides per mouse, 3fields per slide, for a total of 27 fields per cell type). Shown at right arerepresentative photomicrographs of lungs from mice injected with (C)wild-type, (D) a3si, or (E) a3Rx 4T1 cells. Asterisks indicate examples oftumor nodules within the pulmonary parenchyma. H&E, bars ¼ 200 mm.

Figure 4. Experimental lung metastasis is significantly impaired in a3integrin-silenced 4T1 cells. On day 0, 50,000 wild-type, a3si, or a3Rxluciferase-expressing 4T1 cells were injected by tail vein into femaleBalb/Cmice. A, BLI imaging on day 0, immediately after tail vein injectionand on day 14, about halfway through the assay. B, total tumorburden (log photon flux) as measured by BLI. Metastatic colonization bythea3si cells was significantly reduced comparedwithwild-type ora3Rxcells from day 12 onward (�, P < 0.001 vs. wild-type and a3Rx cells,ondays 12–20; #,P <0.05 vs.wild-type andP <0.001 vs.a3Rxonday 22;and �, P < 0.001 vs. a3Rx on day 26, ANOVA with Tukey posttest, n¼ 15mice per group). C, survival to endpoint for mice bearing wild-type, a3si,or a3Rx 4T1 cells. Survival of mice with a3si 4T1 cells was significantlyincreased compared with mice with wild-type or a3Rx 4T1 cells (P <0.0001 vs.wild-type andP¼ 0.0127 vs.a3Rx,Mantel–Cox log-rank test).






our 4T1 cell lines also secrete abundant LM-511, some ofwhich seems to be cell associated (Fig. 7A–F), and some ofwhich is deposited on the substrate (Fig. 7G–I). LM-511secretion was not overtly affected by the loss of a3b1integrin; however, these data confirmed that 4T1 cellsproduce an autocrine extracellular ligand with the potentialto promote growth and survival via a3b1 integrin-depen-dent signaling.To further explore the relationship between laminin secre-

tion and a3b1-dependent proliferation, we silenced a3integrin in EGW593.lu cells, an in vivo-passaged subline ofMDA-MB-231 breast carcinoma cells, which fail to secretedetectable LM-511 or LM-332 (Supplementary Fig. S7A).Despite achieving silencing comparable with or superior tothat which we achieved in 4T1 cells, no obvious differenceswere observed in short-term proliferation assays comparingthe control and a3si MDA-MB-231 cell lines (Supplemen-tary Fig. S7B). Collectively, these data suggested that the roleof a3b1 in breast carcinoma may depend on whether ana3b1 ligand is coexpressed by the carcinoma cells or not.To begin to test the relevance of our findings to human

cancers, we queried TCGA database. Integrin a3 wasupregulated (Z-score > þ2) in 6% of the samples in theBreast Invasive Carcinoma/TCGA Provisional dataset.Among patients with upregulated a3 integrin, those withtumors in which LAMA5 was also upregulated (Z-score >þ1) had significantly worse overall survival than those withtumors in which LAMA5 was not upregulated (mediansurvival time 51 months vs. 114 months; Fig. 7M). Whenexamined separately, changes in mRNA expression of a3integrin or LAMA5 were not on their own linked tosignificant survival differences. The decreased survival wasspecifically linked to cases in which both a3 integrin andLAMA5 were upregulated.

DiscussionIntegrin a3b1 can function to promote breast cancermetastasis in vivoThe ability to evaluate the role of a3 integrin in breast

cancer metastasis had been complicated by the perinatal-lethal phenotype of a3 knockout mice (65). Studies withconditional a3 knockout mice, which were only createdrelatively recently, have thus far shown critical roles for a3integrin in skin tumor formation (66), regulation of mam-mary myoepithelial cell contractility (67), and control ofwound-healing rate (68); however, studies examining theeffect of a3 deletion in breast cancer models have not yetbeen reported. Antibody blockade of a3 can reduce MDA-MB-231 pulmonary arrest (42) and suppress metastaticcolonization of several tumor types upon tail vein, intracar-diac, intracerebral, or peritoneal injection (reviewed inref. 69). However, anti-a3 antibodies have had disparateeffects on tumor cell behaviors in vitro (40, 41), which couldraise questions about the precise mechanism of antibodyaction in vivo. Silencing a3 in MDA-MB-231 cells sup-pressed the growth of tumors implanted both subcutane-ously and inmammary fat pad (43), revealing a role fora3 inpromoting the growth of breast cancer cells at the primarysite, but leaving open the question of the role of a3 inmetastasis.Our new data using the 4T1 cell model now establish that

loss of a3integrin expression can cause a substantial reduc-tion in breast cancer metastatic capacity in vivo. The a3si4T1 cell tumors grew somewhat more slowly at the primarysite, but (i) the reduction of the apparent metastatic tumorburden of the a3si cells was striking compared with themodest reduction in their growth at the primary site, and (ii)analysis of primary tumor size versus lung metastasis did notreveal a correlation, suggesting that the modestly reducedgrowth at the primary site is not sufficient to explain thedramatic reduction in lung metastasis. Our data do not ruleout a contribution by a3b1 to early events in the metastaticcascade, but experimental lung colonization of a3si tumorcells after tail vein injection was reduced to a similar extent asspontaneous metastasis after fat pad injection. This resultsuggests that inhibition of critical, a3-dependent events atthe metastatic site may be sufficient to account for much ofthe reduction in the spontaneous metastasis of a3si cellsfrom the fat pad.

Potential mechanisms for a3b1 integrin contributionsto breast cancer metastasisThe contributions of a3b1 integrin to events at the

metastatic site could include pulmonary arrest, extravasa-tion, and proliferation to form macroscopic metastases.A recent study showed that 4T1 cells selected for enhancedchemotaxis to LM-511 in vitro displayed increased sponta-neous metastasis to bone in vivo, and a snake venomdisintegrin that inhibits a3, a6, and a7 integrins, inhibited4T1 cell migration and invasion toward LM-511 (64).LM-511 is also widely expressed in endothelial cell basementmembranes of the arteries, veins, and capillaries of lung,brain, and lymph nodes (70), and may thus be an important

Figure 6. Impaired proliferation of a3si 4T1 cells. Wild-type, a3si, anda3Rx 4T1 cells were plated in 2% FBS in wells coated with 1 mg/mL LM-332, 1 mg/mL LM-511, 10 mg/mL collagen I, or in uncoated wells. One setof wells was assayed by WST-1 on day 0 (the day of plating) to measurecells input, and the remaining wells were assayed on day 2. Barsrepresent mean� SEM for 6 wells per cell type per condition, normalizedto day 0 values. The growth of the a3si cells was significantly lessthan that of either the wild-type or a3Rx cells on all substrates tested(�, P < 0.001, ANOVA with Tukey posttest).

Zhou et al.





extracellular ligand for tumor cells during extravasation. Weobserved a moderate but significant reduction in a3b1-dependent adhesion and spreading in our a3si cells. Wealso observed reduced proliferation, not only on lamininisoforms, but also on collagen I or on uncoated plastic.This prompted us to examine autocrine production of thea3b1 integrin ligand, LM-511. As previously reported fora bone-metastatic 4T1 cell subclone (14), our 4T1 cellsalso secrete abundant LM-511. Thus, autocrine LM-511may promote 4T1 cell survival and proliferation at themetastatic site, and loss of a3 integrin may partiallydisrupt this autocrine proliferation pathway. In supportof this possibility, silencing a3 integrin in MDA-MB-231–derived EGW593.lu breast carcinoma cells, whichfail to secrete LM-511, did not affect short-term prolif-eration, suggesting that these cells do not depend on such

an autocrine loop. The contributions of a3b1 to MDA-MB-231 tumor growth in vivo may instead involve tumorcell cross-talk to endothelial cells via an a3b1-controlledCOX-2–dependent mechanism (43). Compared with 4T1cells, metastasis of MDA-MB-231 cells occurs with slowerkinetics and fewer metastatic colonies, even thoughimmunocompromised hosts must be used. This suggeststhat coexpression of both a3b1 and an a3b1 ligand inbreast carcinoma may contribute to a more aggressive,metastatic phenotype, a possibility that is supported byour analysis of data from the Breast Invasive CarcinomaTCGA database. The signaling effectors downstream ofa3b1 that support metastasis remain to be defined, butcould include focal adhesion kinase (PTK2), mitogen-activated protein kinase, or Rac1-dependent pathways(71–73).

Figure 7. LAMA5 expression by 4T1cells. A–C, wild-type, a3si, anda3Rx 4T1 cells were cultured oncollagen I-coated coverslipsovernight, then fixed and stainedwith polyclonal anti-LAMA5antibody. D–F, the merged imagesshow counterstaining for F-actinand nuclei with phalloidin andDAPI. G–I, longer exposure imagesin cell-free areas highlight LAMA5deposition on the substrate. J–L,negative control staining withnormal rabbit immunoglobulin.M, survival analysis of a3 integrin-overexpressing human breastcancer cases, based on whether ornot LAMA5 is also upregulated.LAMA5-overexpressing casesshowed reduced survival(P ¼ 0.000403, log-rank test).






Partial loss of a3 function in vitro can produce largeimpacts on metastasis in vivo—a potential therapeuticwindow for a3 integrin antagonists?Compared with the reductions in spontaneous and exper-

imental metastasis that we observed for a3si 4T1 cells invivo, the in vitro phenotypes of the a3si cells were moremodest. It is certainly possible that we have not yet identifiedan in vitro assay that captures the critical function for a3 in4T1 cell metastasis in vivo, but another potential explanationfor this apparent discrepancy could be that there are multiplea3-dependent steps in themetastatic cascade. If three or foursteps in metastasis depend strongly on a3 function, a 40%reduction in a3 function could result in an approximately80% to 90% reduction in metastatic efficiency due tosynergistic effects (0.63 ¼ 0.22, 0.64 ¼ 0.13). In a scenariowhere a3 functions mainly in proliferation at the metastaticsite, it is of interest to note that, in order for a single cell toproduce a macroscopic metastatic colony of approximately1 � 106 cells, approximately 21 cell doublings are required.A reduction in proliferation efficiency of only approximately10% per doubling would then result in around 90% reduc-tion in colony size after 21 doublings (0.921 ¼ 0.11). Thesemathematical considerations suggest that even a modestreduction in a3 function in breast cancer cells could havea dramatic impact on metastatic efficiency. For tumor cellsthat overexpress a3 integrin and critically depend on a3 formetastatic colonization, there may then be a therapeuticwindow wherein interfering with a3 function could signif-icantly curtail metastasis without severely compromising a3function in normal tissues. The ability of a high affinity, a3integrin-binding cyclic peptide to selectively target a3-over-expressing tumor cells in vivo (74, 75), supports the notionthat a3 antagonists might have the potential to specificallytarget tumor cells while leavinga3 function in normal tissuesrelatively intact.In conclusion, our new data now establish that a3

integrin can make critical contributions to breast cancermetastasis in vivo. Seemingly modest reductions in a3function, as measured in in vitro assays, can have asubstantial impact on metastatic behavior in vivo. Thiscould reflect a3 function at multiple levels of the meta-

static cascade or an ongoing requirement for a3 inproliferation at the metastatic site, potentially through anautocrine mechanism in which breast cancer cells takeadvantage of constitutive secretion of an a3 integrinligand, such as LM-511. Future preclinical experimentsshould aim to explore (i) whether a3 antagonists may haveusage as adjuvant therapeutics in combination with otherstrategies to curtail the growth of metastatic a3b1-positivebreast cancers, and (ii) the relationship between thecoexpression of a3b1 and its ligands and breast cancermetastatic behavior.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: B. Zhou, M.E. Herndon, C.S. StippDevelopment of methodology: B. Zhou, K.N Gibson-Corley, M.E. Herndon,Y. Sun, E. Gustafson-Wagner, F.E. DomannAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): B. Zhou, K.N Gibson-Corley, M.E. Herndon, Y. Sun, E. Gustafson-Wagner, M.L. Teoh-Fitzgerald, F.E. Domann, M.D. Henry, C.S. StippAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): B. Zhou, K.N Gibson-Corley, M.E. Herndon, Y. Sun,E. Gustafson-Wagner, C.S. StippWriting, review, and/or revision of the manuscript: B. Zhou, K.N Gibson-Corley,M.E.Herndon, Y. Sun, E.Gustafson-Wagner, F.E.Domann,M.D.Henry, C.S. StippAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): B. Zhou, M.L. Teoh-FitzgeraldStudy supervision: B. Zhou, C.S. Stipp

AcknowledgmentsThe authors thank Jeffrey Miner (Washington University School of Medicine) for

providing anti-laminin a5 antibody. Flow-cytometric data presented herein wereobtained at the Flow Cytometry Facility, which is a Carver College of Medicine CoreResearch Facilities/Holden Comprehensive Cancer Center Core Laboratory at theUniversity of Iowa.

Grant SupportThis work was supported by NIH R01 CA136664 and a seed grant from the

University of Iowa Holden Comprehensive Cancer Center Breast Cancer ResearchInterest Group (to C.S. Stipp), NIH R01 CA115438 (to F.E. Domann), and NIHR01 CA130916 (to M.D. Henry).

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be herebymarked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

Received April 15, 2013; revised July 23, 2013; accepted August 11, 2013;published OnlineFirst September 3, 2013.

References1. Weaver VM, Leli�evre S, Lakins JN, Chrenek MA, Jones JCR, Giancotti

F, et al. beta4 integrin-dependent formation of polarized three-dimen-sional architecture confers resistance to apoptosis in normal andmalignant mammary epithelium. Cancer Cell 2002;2:205–16.

2. Zahir N, Lakins JN, Russell A, Ming W, Chatterjee C, Rozenberg GI,et al. Autocrine laminin-5 ligates alpha6beta4 integrin and activatesRAC and NFkappaB to mediate anchorage-independent survival ofmammary tumors. J Cell Biol 2003;163:1397–407.

3. Plopper GE, Domanico SZ, Cirulli V, Kiosses WB, Quaranta V. Migra-tion of breast epithelial cells on Laminin-5: differential role of integrinsin normal and transformed cell types. Breast Cancer Res Treat 1998;51:57–69.

4. MartinKJ,KwanCP,Nagasaki K, ZhangX,O'HareMJ,KaelinCM, et al.Down-regulation of laminin-5 in breast carcinoma cells. Mol Med1998;4:602–13.

5. Henning K, Berndt A, Katenkamp D, Kosmehl H. Loss of laminin-5 inthe epithelium-stroma interface: an immunohistochemical marker ofmalignancy in epithelial lesions of the breast. Histopathology 1999;34:305–9.

6. Sathyanarayana UG, Padar A, Huang CX, Suzuki M, Shigematsu H,Bekele BN, et al. Aberrant promoter methylation and silencing oflaminin-5-encoding genes in breast carcinoma. Clin Cancer Res2003;9:6389–94.

7. Korah R, Das K, Lindy ME, Hameed M, Wieder R. Coordinate loss offibroblast growth factor 2 and laminin 5 expression during neoplas-tic progression of mammary duct epithelium. Hum Pathol 2007;38:154–60.

8. Carpenter PM,Wang-Rodriguez J, ChanOTM,Wilczynski SP. Laminin5 expression in metaplastic breast carcinomas. Am J Surg Pathol2008;32:345–53.

Zhou et al.





9. Carpenter PM,DaoAV, Arain ZS,ChangMK,NguyenHP,Arain S, et al.Motility induction in breast carcinoma by mammary epithelial laminin332 (laminin 5). Mol Cancer Res 2009;7:462–75.

10. Kwon S-Y, Chae SW, Wilczynski SP, Arain A, Carpenter, Philip M.Laminin 332 expression in breast carcinoma. Appl ImmunohistochemMol Morphol 2012;20:159–64.

11. KimBG, AnHJ, KangS, Choi YP, GaoM-Q, Park H, et al. Laminin-332-rich tumormicroenvironment for tumor invasion in the interface zone ofbreast cancer. Am J Pathol 2011;178:373–81.

12. Hewitt RE, Powe DG, Morrell K, Balley E, Leach IH, Ellis IO, et al.Laminin and collagen IV subunit distribution in normal and neoplastictissues of colorectum and breast. Br J Cancer 1997;75:221–9.

13. M€a€att€a M, Virtanen I, Burgeson R, Autio-Harmainen H. Comparativeanalysis of the distribution of laminin chains in the basement mem-branes in somemalignant epithelial tumors: the alpha1 chain of lamininshows a selected expression pattern in human carcinomas. J Histo-chem Cytochem 2001;49:711–26.

14. Chia J, Kusuma N, Anderson R, Parker B, Bidwell B, Zamurs L, et al.Evidence for a role of tumor-derived laminin-511 in the metastaticprogression of breast cancer. Am J Pathol 2007;170:2135–48.

15. Pouliot N, KusumaN. Laminin-511: amulti-functional adhesionproteinregulating cell migration, tumor invasion andmetastasis. Cell AdhMigr2013;7:142–9.

16. Siler U, Seiffert M, Puch S, Richards A, Torok-Storb B, M€uller CA, et al.Characterization and functional analysis of laminin isoforms in humanbone marrow. Blood 2000;96:4194–203.

17. Gu Y, Sorokin L, Durbeej M, Hjalt T, J€onsson JI, Ekblom M. Charac-terization of bone marrow laminins and identification of alpha5-con-taining laminins as adhesive proteins for multipotent hematopoieticFDCP-Mix cells. Blood 1999;93:2533–42.

18. Pierce RA, Griffin GL, Miner JH, Senior RM. Expression patterns oflaminin alpha1 and alpha5 in human lung during development. Am JRespir Cell Mol Biol 2000;23:742–7.

19. Diaz LK, Cristofanilli M, Zhou X, Welch KL, Smith TL, Yang Y, et al.Beta4 integrin subunit gene expression correlates with tumor sizeand nuclear grade in early breast cancer. Mod Pathol 2005;18:1165–75.

20. Lu S, Simin K, Khan A, Mercurio AM. Analysis of integrin beta4expression in human breast cancer: association with basal-liketumors and prognostic significance. Clin Cancer Res 2008;14:1050–8.

21. Lipscomb EA, Mercurio AM. Mobilization and activation of a signalingcompetent alpha6beta4integrin underlies its contribution to carcinomaprogression. Cancer Metastasis Rev 2005;24:413–23.

22. Wilhelmsen K, Litjens SHM, Sonnenberg A. Multiple functions of theintegrin alpha6beta4 in epidermal homeostasis and tumorigenesis.Mol Cell Biol 2006;26:2877–86.

23. Giancotti FG. Targeting integrin beta4 for cancer and anti-angiogenictherapy. Trends Pharmacol Sci 2007;28:506–11.

24. Shaw LM, Rabinovitz I, Wang HH, Toker A, Mercurio AM. Activation ofphosphoinositide 3-OH kinase by the alpha6beta4 integrin promotescarcinoma invasion. Cell 1997;91:949–60.

25. ShawLM. Identification of insulin receptor substrate 1 (IRS-1) and IRS-2 as signaling intermediates in the alpha6beta4 integrin-dependentactivation of phosphoinositide 3-OHkinase andpromotion of invasion.Mol Cell Biol 2001;21:5082–93.

26. O'Connor KL, Shaw LM, Mercurio AM. Release of cAMP gating bythe alpha6beta4 integrin stimulates lamellae formation and the che-motactic migration of invasive carcinoma cells. J Cell Biol 1998;143:1749–60.

27. O'Connor KL, Nguyen BK, Mercurio AM. RhoA function in lamellaeformation and migration is regulated by the alpha6beta4 integrin andcAMP metabolism. J Cell Biol 2000;148:253–8.

28. O'Connor KL, ChenM, Towers LN. Integrin a6b4 cooperates with LPAsignaling to stimulate Rac through AKAP-Lbc-mediated RhoA activa-tion. Am J Physiol Cell Physiol 2012;302:C605–14.

29. Chung J, Bachelder RE, Lipscomb EA, Shaw LM, Mercurio AM.Integrin (alpha 6 beta 4) regulation of eIF-4E activity and VEGF trans-lation: a survival mechanism for carcinoma cells. J Cell Biol 2002;158:165–74.

30. Bertotti A, Comoglio PM, Trusolino L. Beta4 integrin activates a Shp2-Src signaling pathway that sustains HGF-induced anchorage-inde-pendent growth. J Cell Biol 2006;175:993–1003.

31. Bertotti A, Comoglio PM, Trusolino L. Beta4 integrin is a transformingmolecule that unleashes Met tyrosine kinase tumorigenesis. CancerRes 2005;65:10674–9.

32. GuoW, Pylayeva Y, Pepe A, Yoshioka T, Muller WJ, Inghirami G, et al.Beta 4 integrin amplifies ErbB2 signaling to promote mammary tumor-igenesis. Cell 2006;126:489–502.

33. Dutta U, Shaw LM. A key tyrosine (Y1494) in the beta4 integrinregulates multiple signaling pathways important for tumor develop-ment and progression. Cancer Res 2008;68:8779–87.

34. Yang X, Dutta U, Shaw LM. SHP2 mediates the localized activation ofFyn downstream of the a6b4 integrin to promote carcinoma invasion.Mol Cell Biol 2010;30:5306–17.

35. Merdek KD, Yang X, Taglienti CA, Shaw LM, Mercurio AM. Intrinsicsignaling functions of the beta4 integrin intracellular domain. J BiolChem 2007;282:30322–30.

36. Gui GP, Wells CA, Yeomans P, Jordan SE, Vinson GP, Carpenter R.Integrin expression in breast cancer cytology: a novel predictor ofaxillary metastasis. Eur J Surg Oncol 1996;22:254–8.

37. Gui GP, Wells CA, Browne PD, Yeomans P, Jordan S, Puddefoot JR,et al. Integrin expression in primary breast cancer and its relation toaxillary nodal status. Surgery 1995;117:102–8.

38. Tawil NJ, Gowri V, Djoneidi M, Nip J, Carbonetto S, Brodt P. Integrinalpha3beta1 can promote adhesion and spreading of metastaticbreast carcinoma cells on the lymph node stroma. Int J Cancer1996;66:703–10.

39. Lundstr€om A, Holmbom J, Lindqvist C, Nordstr€om T. The roleof alpha2 beta1 and alpha3 beta1 integrin receptors in the initialanchoring of MDA-MB-231 human breast cancer cells to corticalbone matrix. Biochem Biophys Res Commun 1998;250:735–40.

40. Sugiura T, Berditchevski F. Function of alpha3beta1-tetraspanin pro-tein complexes in tumor cell invasion. Evidence for the role of thecomplexes in production ofmatrixmetalloproteinase 2 (MMP-2). J CellBiol 1999;146:1375–89.

41. Morini M, Mottolese M, Ferrari N, Ghiorzo F, Buglioni S, Mortarini R,et al. The alpha 3 beta 1 integrin is associated with mammary carci-noma cellmetastasis, invasion, and gelatinaseB (MMP-9) activity. Int JCancer 2000;87:336–42.

42. Wang H, Fu W, Im JH, Zhou Z, Santoro SA, Iyer V, et al. Tumor cellalpha3beta1 integrin and vascular laminin-5 mediate pulmonary arrestand metastasis. J Cell Biol 2004;164:935–41.

43. Mitchell K, Svenson KB, Longmate WM, Gkirtzimanaki K, Sadej R,Wang X, et al. Suppression of integrin alpha3beta1 in breast cancercells reduces cyclooxygenase-2 gene expression and inhibits tumor-igenesis, invasion, and cross-talk to endothelial cells. Cancer Res2010;70:6359–67.

44. Varzavand A, Drake JM, Svensson RU, Herndon ME, Zhou B, HenryMD, et al. Integrin a3b1 regulates tumor cell responses to stromal cellsand can function to suppress prostate cancer metastatic colonization.Clin Exp Metastasis 2013;30:541–52.

45. Bao L, Haque A, Jackson K, Hazari S,Moroz K, Jetly R, et al. Increasedexpression of P-glycoprotein is associatedwith doxorubicin chemore-sistance in the metastatic 4T1 breast cancer model. Am J Pathol2011;178:838–52.

46. Zevian S, Winterwood NE, Stipp CS. Structure-function analysis oftetraspanin CD151 reveals distinct requirements for tumor cell beha-viors mediated by a3b1 versus a6b4 integrin. J Biol Chem 2011;286:7496–506.

47. Miner JH, Patton BL, Lentz SI, Gilbert DJ, SniderWD, JenkinsNA, et al.The laminin alpha chains: expression, developmental transitions, andchromosomal locations of alpha1-5, identification of heterotrimericlaminins 8–11, and cloning of a novel alpha3 isoform. J Cell Biol1997;137:685–701.

48. Winterwood NE, Varzavand A, Meland MN, Ashman LK, Stipp CS. Acritical role for tetraspanin CD151 in alpha3beta1 and alpha6beta4integrin-dependent tumor cell functions on laminin-5. Mol Biol Cell2006;17:2707–21.






49. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25years of image analysis. Nat Methods 2012;9:671–5.

50. Stipp CS, Kolesnikova TV, Hemler ME. EWI-2 regulates alpha3beta1integrin-dependent cell functions on laminin-5. J Cell Biol 2003;163:1167–77.

51. Drake JM, Gabriel CL, Henry MD. Assessing tumor growth and dis-tribution in a model of prostate cancer metastasis using biolumines-cence imaging. Clin Exp Metastasis 2005;22:674–84.

52. Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al.The cBio cancer genomics portal: an open platform for exploringmultidimensional cancer genomics data. Cancer Discov 2012;2:401–4.

53. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al.Integrative analysis of complex cancer genomics and clinical profilesusing the cBioPortal. Sci Signal 2013;6:pl1.

54. Black PC, Shetty A, Brown GA, Esparza-Coss E, Metwalli AR, AgarwalPK, et al. Validating bladder cancer xenograft bioluminescence withmagnetic resonance imaging: the significance of hypoxia and necro-sis. BJU Int 2010;106:1799–804.

55. Hilali El N, Rubio N, Martinez-Villacampa M, Blanco J. Combinednoninvasive imaging and luminometric quantification of luciferase-labeled human prostate tumors and metastases. Lab Invest 2002;82:1563–71.

56. Hakem A, Sanchez-Sweatman O, You-Ten A, Duncan G, Wakeham A,Khokha R, et al. RhoC is dispensable for embryogenesis and tumorinitiation but essential for metastasis. Genes Dev 2005;19:1974–9.

57. Rosenthal DT, Zhang J, Bao L, Zhu L,Wu Z, Toy K, et al. RhoC impactsthe metastatic potential and abundance of breast cancer stem cells.PLoS ONE 2012;7:e40979.

58. Yang J, Mani SA, Donaher JL, Ramaswamy S, Itzykson RA, Come C,et al. Twist, a master regulator of morphogenesis, plays an essentialrole in tumor metastasis. Cell 2004;117:927–39.

59. Ma L, Reinhardt F, Pan E, Soutschek J, Bhat B, Marcusson EG, et al.Therapeutic silencing of miR-10b inhibits metastasis in a mousemammary tumor model. Nat Biotechnol 2010;28:341–7.

60. Bourguignon LYW, Wong G, Earle C, Krueger K, Spevak CC. Hyalur-onan-CD44 interaction promotes c-Src-mediated twist signaling,microRNA-10b expression, and RhoA/RhoC up-regulation, leading toRho-kinase-associated cytoskeleton activation and breast tumor cellinvasion. J Biol Chem 2010;285:36721–35.

61. Granot Z, Henke E, Comen EA, King TA, Norton L, Benezra R. Tumorentrained neutrophils inhibit seeding in the premetastatic lung. CancerCell 2011;20:300–14.

62. Subbaram S, Dipersio CM. Integrin a3b1 as a breast cancer target.Expert Opin Ther Targets 2011;15:1197–210.

63. Kusuma N, Anderson RL, Pouliot N. Laminin a5-derived peptidesmodulate the properties of metastatic breast tumour cells. Clin ExpMetastasis 2011;28:909–21.

64. KusumaN, Denoyer D, Eble JA, Redvers RP, Parker BS, Pelzer R, et al.Integrin-dependent response to laminin-511 regulates breast tumorcell invasion and metastasis. Int J Cancer 2012;130:555–66.

65. Kreidberg JA, Donovan MJ, Goldstein SL, Rennke H, Shepherd K,Jones RC, et al. Alpha 3 beta 1 integrin has a crucial role in kidney andlung organogenesis. Development 1996;122:3537–47.

66. Sachs N, Secades P, van Hulst L, Kreft M, Song J-Y, Sonnenberg A.Loss of integrin a3 prevents skin tumor formation by promotingepidermal turnover and depletion of slow-cycling cells. Proc Natl AcadSci USA 2012;109:21468–73.

67. Raymond K, Cagnet S, Kreft M, Janssen H, Sonnenberg A, GlukhovaMA. Control of mammary myoepithelial cell contractile function bya3b1 integrin signalling. EMBO J 2011;30:1896–906.

68. Margadant C, Raymond K, Kreft M, Sachs N, Janssen H, SonnenbergA. Integrin {alpha}3{beta}1 inhibits directional migration andwound re-epithelialization in the skin. J Cell Sci 2009;122:278–88.

69. Stipp CS. Laminin-binding integrins and their tetraspanin partners aspotential antimetastatic targets. Expert Rev Mol Med 2010;12:e3.

70. Sorokin LM, Pausch F, Frieser M, Kr€oger S, Ohage E, Deutzmann R.Developmental regulation of the laminin alpha5 chain suggests arole in epithelial and endothelial cell maturation. Dev Biol 1997;189:285–300.

71. Manohar A, Shome SG, Lamar J, Stirling L, Iyer V, Pumiglia K, et al.Alpha 3 beta 1 integrin promotes keratinocyte cell survival throughactivation of a MEK/ERK signaling pathway. J Cell Sci 2004;117:4043–54.

72. Choma DP, Pumiglia K, Dipersio CM. Integrin alpha3beta1 directs thestabilization of a polarized lamellipodium in epithelial cells throughactivation of Rac1. J Cell Sci 2004;117:3947–59.

73. ChomaDP,MilanoV,Pumiglia KM,DipersioCM. Integrin alpha3beta1-dependent activation of FAK/Src regulates Rac1-mediated keratino-cyte polarization on laminin-5. J Invest Dermatol 2007;127:31–40.

74. Aina OH, Liu R, Sutcliffe JL, Marik J, Pan C-X, Lam KS. From com-binatorial chemistry to cancer-targeting peptides. Mol Pharm 2007;4:631–51.

75. Xiao K, Li Y, Lee JS, Gonik AM, Dong T, Fung G, et al. "OA02" peptidefacilitates the precise targeting of paclitaxel-loaded micellar nanopar-ticles to ovarian cancer in vivo. Cancer Res 2012;72:2100–10.

Zhou et al.





2014;12:143-154. Published OnlineFirst September 3, 2013.Mol Cancer Res Bo Zhou, Katherine N. Gibson-Corley, Mary E. Herndon, et al. and Lung Colonization of Invasive Breast Carcinoma

1 Can Function to Promote Spontaneous Metastasisβ3αIntegrin

Updated version

10.1158/1541-7786.MCR-13-0184doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2013/09/05/1541-7786.MCR-13-0184.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/12/1/143.full#ref-list-1

This article cites 75 articles, 34 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/12/1/143.full#related-urls

This article has been cited by 5 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mcr.aacrjournals.org/content/12/1/143To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-13-0184

http://mcr.aacrjournals.org/content/suppl/2013/09/05/1541-7786.MCR-13-0184.DC1

http://mcr.aacrjournals.org/content/12/1/143.full#ref-list-1

http://mcr.aacrjournals.org/content/12/1/143.full#related-urls

http://mcr.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mcr.aacrjournals.org/content/12/1/143


integrin a3b1 can function to promote spontaneous ... · signal transduction integrin a3b1 can...

Documents